PROJECT SUMMARY/ABSTRACT There exist very few effective treatments that ease the intelligibility burden of dysarthria, and all of these require cognitive and physical effort on the part of the speaker to achieve and maintain gains [1,2]. Therefore, individuals with intelligibility deficits whose cognitive and physical impairments limit their ability to modify their speech are simply not viable treatment candidates. This constitutes a significant health disparity that disproportionately affects those populations with developmental, cognitive, and/or significant neuromuscular impairment. This proposal addresses this critical gap in clinical practice, by shifting the weight of behavioral change from the speaker to the listener. While a novel concept for dysarthria management, the idea is firmly rooted in the field of psycholinguistics and supported by our programmatic body of research showing that listener-targeted perceptual training paradigms (wherein listeners are familiarized with the degraded speech signal and provided with an orthographic transcription of what the speaker is saying) result in statistically and clinically significant intelligibility gains in dysarthria [5-14]. Further, we have preliminary evidence to suggest that these intelligibility outcomes may be influenced by hypothesis-driven speaker parameters, such as acoustic predictability of speech rhythm cues [5,15], and listener parameters, such as expertise in rhythm perception [5,6]. A requisite next step to bringing listener-targeted perceptual training closer to clinical implementation, and the overarching goal of the work proposed herein, is the systematic and rigorous analysis of the speaker and listener parameters, and their interactions, that modulate, and in some cases optimize, perceptual training benefits of intelligibility improvement. To achieve this aim, we utilize an existing database of dysarthric speech and a large cohort of listeners across two well-established testing sites. The key deliverable from this proposal will be explanatory models that account for the unique and joint contributions of speaker and listener parameters on the magnitude of intelligibility improvement following perceptual training with dysarthric speech. Importantly, the explanatory models derived from this work will not only inform clinical decision-making regarding candidacy for this potential treatment option, but also will be leveraged in a follow-on Phase I Clinical Trial to identify training task-related parameters (e.g., dose, feedback, procedures) that optimize both generalization and maintenance of intelligibility outcomes associated with perceptual training with dysarthric speech. Perceptual training offers a promising avenue for improving intelligibility of dysarthric speech by offsetting the communicative burden from the patient with dysarthria on to their primary communication partners?family, friends, and caregivers. Thus, this work sets the stage for an innovative and much needed extension of dysarthria management to listener-targeted remediation?laying the groundwork for a speech intelligibility intervention option for individuals for whom one currently does not exist.